547 Background: Rebeccamycin analog (NSC 655649) is a synthetic antibiotic cytotoxic agent thought to inhibit topoisomerase function. We sought to determine the response rate to rebeccamycin analog among patients with refractory advanced breast cancer using two different treatment schedules. Methods: Eligible patients had advanced breast cancer and measurable tumor (RECIST), and 1 or 2 prior chemotherapy regimens for advanced breast cancer, or recurrence within 12 months of adjuvant chemotherapy. Central venous access was required, as were adequate bone marrow, hepatic, and renal reserve. Patients were randomized to receive rebeccamycin analog on one of two treatment schedules: arm 1, 500 mg/m2 IV bolus every 21 days; arm 2, 140 mg/m2 IV bolus daily x 5 days, every 21 days. The primary study endpoint was response rate. A two stage accrual design was used to evaluate each schedule separately. At least 3 responses among 21 patients in either arm were required to reject the null hypothesis of a 5% response rate for the alternative of a 20% response rate. Results: Forty-two women entered the trial (median age 54.7 yrs), 21 on each arm. Prior chemotherapy regimens for metastastic breast cancer were: 0, n=4, 1, n=21; 2, n=17. Prior treatments (including adjuvant therapy) anthracyclines: 90%, taxanes 64%, 5FU-based therapy, 48%. There were 4 partial responses (overall response rate 9.5%), two in each treatment arm. Median time to progression was 2.1 months (range 1–11+ months). All responding patients had received anthracycline-based adjuvant chemotherapy and at least one prior regimen for metastatic breast cancer. An additional 10 patients had stable disease as best response, including 4 with SD > 6 months. Treatment was generally well tolerated. Grade 3 or 4 toxicity rates were: anemia 5%, neutropenia 33%, thrombocytopenia 12%, RBC transfusion 14%, nausea/vomiting 10%. Other side effects were very mild. Toxicity profiles were similar between the two treatment arms. Conclusions: Rebeccamycin analog is reasonably well tolerated on two different treatment schedules, but had modest clinical activity in this heavily pretreated patient population. No significant financial relationships to disclose.